Construction and Building Materials 148 (2017) 675–685

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Accelerated corrosion of pipeline steel and reduced cathodic protection

effectiveness under direct current interference
Shan Qian, Y. Frank Cheng ⇑
Department of Mechanical Engineering, University of Calgary, Calgary, Alberta T2N 1N4, Canada

h i g h l i g h t s

Measured the potential of pipeline steel under direct current (DC) densities.

Quantified the accelerated corrosion rate of the steel under DC interference.

Determined the shift of cathodic protection potential in the presence of DC.

Determined the electrolyte alkalinisation caused by DC stray current.

a r t i c l e i n f o a b s t r a c t

Article history: In this work, corrosion of an X52 pipeline steel at various direct current (DC) densities was investigated in
Received 6 February 2017 an extracted soil solution by potential and pH measurements, weight-loss testing and optical microscopy.
Received in revised form 19 April 2017 It was found that the DC stray current could polarize, either anodically or cathodically, the steel at the
Accepted 5 May 2017
anodic and cathodic zones, respectively, resulting in accelerated corrosion of the steel and the cathodic
Available online 17 May 2017
reduction of dissolved oxygen. The CP potential was not maintained at the applied value under the DC
interference, and was shifted to positive and negative directions in the anodic and cathodic zones, respec-
Keywords:
tively. The pipeline suffered from both anodic and cathodic DC interferences. The steel in the anodic zone
Pipeline steel
Corrosion
cannot be protected, at least not fully protected, by the CP from corrosion. The solution alkalization in the
Direct current interference cathodic zone would be further enhanced due to the combined effect of DC and CP on cathodic reaction.
Cathodic protection The implications of the DC interference on pipeline integrity were discussed.
Solution pH Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction
Stray current refers to the electric current flowing through con-
ductive paths other than the intended ones [1]. Corrosion of metals
can be caused by the stray current interference [2]. In the 1960s,
Schwalm and Sandor [3] determined that the stray current, includ-
ing both alternating current (AC) and direct current (DC) ones, was
able to accelerate corrosion of buried pipelines. The stray current
can be categorized into dynamic or static ones based on the fact
whether the amplitude and direction of the current change over
time [4,5]. Generally, the static stray current interference that a
pipeline is subject to is generated by outside current sources, such
as high-voltage AC or DC powerlines, DC rectifiers, etc. The
dynamic current interference is mostly from the orbital transport
system, i.e., city transit systems. When collocated with the stray
⇑ Corresponding author.
E-mail address: fcheng@ucalgary.ca (Y.F. Cheng).
current sources, the pipeline would suffer from accelerated corro-
sion, resulting in the so-called AC or DC corrosion [6–15].
The DC corrosion of buried pipelines is similar to classic corro-
sion mechanism, i.e., current flows from anode to cathode through
the soil electrolyte by ionic conductivity and from cathode to
anode through the pipeline by electrons [16]. At the anode, the
pipe steel oxidizes; and at the cathode, hydrogen evolution or
the oxygen reduction occurs. Thus, steel corrosion occurs at the
anode, with no corrosion attack at the cathode. The corrosion
kinetics obeys Faraday’s law, i.e., 1 ampere of DC discharge from
a pipeline may consume 10 kg of steel within one year [17]. Gener-
ally, the DC can cause more serious corrosion than AC on pipeline
steels. McIntosh [18] demonstrated that the corrosion of steels
under AC interference of 60 Hz is about 1% of that induced by an
equivalent amount of DC. Moreover, the DC corrosion is localized
in nature. Corrosion reaction occurs at the anode where the pipe
steel is exposed to soil due to local breakage of external coatings,
which can result in dramatic consequences on pipelines [19,20].

http://dx.doi.org/10.1016/j.conbuildmat.2017.05.024
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
676 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685
Different from AC corrosion, there have been limited investiga-
tions on DC corrosion of pipelines, including parametric effects and
the mutual interference of DC with cathodic protection (CP) [21–
24]. For example, the CP criterion recommended by NACE Interna-
tional to protect underground steel structures from corrosion
attack is a cathodic potential of 850 mV vs. copper sulfate elec-
trode (CSE) [25]. However, the presence of DC interference can
decrease the effectiveness of CP by shifting the applied CP potential
from the designed value. There has been so far rare understanding
of the quantitative relationship between DC current density and
the CP potential shift.
The DC corrosion of buried pipelines has attracted increasing
attention recently along with the design and construction of
high-voltage DC (HVDC) power lines, which are usually located
in energy corridors where oil/gas pipelines are buried. This is par-
ticularly true in Alberta, Canada, where the proposed HVDC lines
are adjacent to energy pipelines over a long distance. Obviously,
investigations of DC corrosion of pipelines not only improve our
understanding to this important phenomenon, but also meet
immediate industry and society needs for safe and reliable design
of energy infrastructure.
In this work, corrosion of an X52 pipeline steel at various DC
current densities was investigated in an extracted soil solution
by potential and pH measurements, weight-loss testing and optical
microscopy. The effect of DC current density on both anodic and
cathodic reactions was determined. The shift of CP potential under
DC interference was measured. The mechanistic aspects about the
accelerated pipeline corrosion and reduced CP effectiveness in the
presence of DC interference were analyzed. The implications of the
DC interference on pipeline integrity were discussed.
2. Experimental
2.1. Specimen and solution
Specimens used in this work were cut from an X52 steel pipe, with a chemical
composition (wt.%): 0.24 C, 1.4 Mn, 0.45 Si, 0.025 P, 0.015 S, 0.10 V, 0.05 Nb, 0.04 Ti
and Fe balance. The steel coupons were machined into 10 mm  10 mm, and were
sealed with epoxy, leaving an area of 100 mm2. The specimen preparation was con-
trolled carefully to ensure that there was no bubble and crevice at the epoxy/steel
interface. All specimens were ground with 240, 400, 600, 800, and 1000 grit emery
papers, followed by cleaning in distilled water and methanol, and drying in air.
To simulate the Regina clay soil (Saskatoon, Canada) where the pipeline was
buried, an extracted soil solution was used in this work, which contained
0.0755 g/L NaHCO3, 0.0092 g/L NaCl, 0.0014 g/L NaNO3, 0.0773 g/L Na2SO4,
0.0619 g/L K2SO4, 1.116 g/L CaSO42H2O and 0.662 g/L MgSO47H2O. The pH of the
extracted soil solution was 7.6. The solution was made from analytic grade reagents
and ultra-pure water. All tests were conducted at room temperature of about 22 °C.
2.2. Electrochemical measurements
Fig. 1 shows the home-designed experimental setup for DC corrosion testing.
There were two cells, i.e., anodic cell and cathodic cell, to simulate the anodic
and cathodic zones during DC corrosion of pipelines. Two different electric circuits,
i.e., DC stray current circuit and CP circuit, were included in the setup, where the DC
system was under galvanostatic control and the CP system was under potentiostatic
control. In the DC current circuit, two carbon rods, which were used as counter elec-
trodes (CE) in the anodic and cathodic cells, were connected to the negative and
positive poles of a DC power supply, respectively. The steel specimens, which were
used as working electrodes (WE) in both cells, were connected through a slide rheo-
stat to form a closed stray DC circuit. The DC current density flowing between WE
and CE was varied, i.e., 0, 0.1, 0.5, 1, 2, 5 and 10 A/m2, by adjusting the voltage out-
put of the power supply. As the DC output was under control galvanostatically, the
DC current density would maintain constantly in order to investigate the effect of
DC current density on pipeline corrosion. It was realized that, in reality, the change
of soil conductivity will change the DC current density flowing on the buried pipe-
line, and thus the DC corrosion.
In the CP circuit, the steel specimen and a carbon rod were connected to the
negative and positive poles of another DC power supply, respectively. A saturated
calomel electrode (SCE) was used as reference electrode (RE), which was immersed
into a Luggin capillary. The distance from the tip of the capillary to the steel elec-
trode surface was 2 mm in order to eliminate the ohmic drop. It was noted that
the potential of the steel cannot be maintained at the applied CP potential when
the DC circuit was ON. In other words, the potential would deviate from the applied
CP potential under the DC effect (i.e., the potential of the steel did not automatically
adjust itself to the applied CP potential). The potential of the steel electrode under
various DC densities was measured through a Gamry Reference 600 electrochemical
system. The recorded potential of the steel included both the applied CP potential
and the potential shift under DC stray current. All potentials in this work were con-
verted to CSE to be consistent with the industry use.
During testing, the test cells were continuously purged with air to maintain the
constant dissolved oxygen content in the solution. A dissolved oxygen sensor was
used to monitor the content of dissolved oxygen, which was 7.7 ppm. A pH meter
(Oaklon Acorn) was installed to monitor the solution pH at the anodic and cathodic
cells. To ensure the reproducibility of the measured data, each test was conducted
three time.
2.3. Weight-loss testing
Prior to weight-loss testing, steel coupons were weighed by an electronic bal-
ance with an accuracy of 0.1 mg. They were immersed in the soil solution under
the corrosion potential of the steel and two CP potentials, i.e., 850 mV vs. CSE
and 1000 mV vs. CSE, respectively, and varied DC current densities. To ensure
reproducibility of the testing results, three parallel coupons were used at each
condition.
After testing, the corrosion products formed on the steel surface were removed
carefully by both mechanical and chemical methods according to ASTM G1-03 [26].
Light scraping and scrubbing was used to remove tightly adherent corrosion prod-
ucts, and a descaling solution containing 500 mL HCl, 3.5 g hexamethylenete-
tramine and 500 mL distilled water was used in chemical removal procedure. The
cleaning process was repeated several times to remove corrosion products thor-
oughly. After rinsing with distilled water and alcohol, the steel coupons were dried
and weighed again. The weight change under specific testing conditions was
recorded. The corrosion rate (mm/year) of the steel was calculated from the
weight-loss measurements and the testing time.
2.4. Surface characterization
After 48 h of testing in the soil solution at various DC current densities and/or
CP potentials, the steel electrodes were removed from the solution, and rinsed with
deionized water. The surface morphology of the steel was observed by an optical
microscope.
3. Results and discussion
3.1. Potentials of the steel in the soil solution under various DC current
densities
Fig. 2 shows the time dependence of the potential of X52 steel
electrode in the anodic zone at various DC current densities in
the soil solution. In all recorded curves, three potential regions
are observed. From 0 to 1200 s, there is no DC applied. The
recorded potential is the corrosion potential of the steel in the
solution. From 1200 to 2400 s, the DC is applied and the DC-ON
potential is recorded. After 2400 s, the DC is stopped and the DC-
OFF potential is recorded. It is seen that, upon application of DC,
the potential of the steel in the anodic zone is shifted less nega-
tively (i.e., towards the positive direction). The potential becomes
less negative with the increasing DC current density. At each DC
current density, the potential of the steel is maintained at a relative
stable value. When the DC is stopped at the 2400th s, the potential
drops rapidly, and then gradually reaches a steady state value,
which is approximately identical to the corrosion potential
recorded prior to the DC application, i.e., the potential from 0 to
1200 s. At higher DC current densities, the potential recovery to
the steady value is a bit slower.
Fig. 3 shows the time dependence of the potential of X52 steel
electrode under various DC densities in the soil solution in the
cathodic zone. Similarly, before the DC is applied, the steel is in
its corrosion potential of about 800 mV vs. CSE. Upon application
of the DC current density, the potential of the steel is shifted neg-
atively and remains at a constant value. As the increasing DC cur-
rent density, the potential tends to be more negative. When the DC
is stopped at the 2400th s, the DC potential is shifted towards the
positive direction, but does not return to the corrosion potential
 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685 677

Fig. 1. Schematic diagram of the experimental setup for DC corrosion testing on X52 steel specimen in the extracted soil solution, where WE, RE and CE refer to working
electrode, reference electrode and counter electrode, respectively, and SCE is saturated calomel electrode. The DC parameters to be tested are also included.

recorded before the DC is applied. The new potential is generally
less negative than the initial corrosion potential in the period from
0 to 1200 s.
Fig. 4 shows the steady-state DC-ON potential of the X52 steel
electrode in anodic and cathodic zones, respectively, in the soil
solution as a function of DC current density. It is seen that, in the
anodic zone, the potential becomes less negative with the increas-
ing DC current density; while in the cathodic zone, the potential is
more negative at higher DC current densities.
3.2. Solution pH at anodic and cathodic zones under various DC
various densities
Fig. 5 shows the solution pH in the anodic and cathodic zones,
respectively, as a function of DC current density. It is seen that,
in the anodic zone, the solution pH drops (tends to be more acidic)
with the increasing DC current density (Fig. 5a). However, the
solution pH in the cathodic zone increases (tends to be more alka-
line) with the increase of the DC current density (Fig. 5b).
3.3. Corrosion rate and morphology of the steel under various DC
current densities
Fig. 6 shows the corrosion rate determined by weight-loss test-
ing of X52 steel after 48 h of immersion in the soil solution as a
function of DC current density. It is seen that the corrosion rate
of the steel increases with the DC current density. For example,
in the absence of DC, the steel corrodes in the aerated soil solution
at a corrosion rate of about 0.4 mm/y. At a DC current density of 5
A/m2 and 10 A/m2, the corrosion rate is increased to 6.5 mm/year
and 12.5 mm/year, respectively.
Fig. 7 shows the optical views of the morphology of the steel
electrodes after 48 h of testing in the soil solution under various
DC current densities. In the absence of DC, the steel electrode
678 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685

-0.2 -0.3
0.1 A/m2 (a)
0.5 A/m2
10 A/m2 -0.4
-0.4 1 A/m2
2 A/m2

Potential (V vs. CSE)

Potential (V vs. CSE)

5 A/m2
5 A/m2
-0.6 -0.5
2 A/m2 10 A/m2
1 A/m2

0.5 A/m2
-0.8 0.1 A/m2
-0.6

Corrosion DC-ON DC-OFF -0.7

-1.0
potential potential potential
-0.8
-1.2

-0.9
-1.4 0 2 4 6 8 10 12
0 500 1000 1500 2000 2500 3000 3500 4000
DC current density (A/m2)
Time (s)
-0.7
Fig. 2. Time dependence of the potential of X52 steel electrode in the anodic zone at
various DC current densities in the soil solution. (b)
-0.8

-0.9

Potential (V vs. CSE)

0.0
-1.0
-0.2
-1.1
Potential (V vs. CSE)

-0.4 Corrosion DC-ON DC-OFF -1.2

potential potential potential
-0.6 -1.3

-0.8 0.1 A/m2

-1.4

-1.0 0.1 A/m2 -1.5

0.5 A/m2 1 A/m2
0.5 A/m2
-1.2 1 A/m2 -1.6
2 A/m2 0 2 4 6 8 10 12
2 A/m2
5 A/m2 5 A/m2 2
-1.4 DC current density (A/m )
10 A/m2
10 A/m2
-1.6 Fig. 4. Steady-state DC-ON potential of the X52 steel electrode in (a) anodic zone
0 500 1000 1500 2000 2500 3000 3500 4000 and (b) cathodic zone as a function of DC current density in the soil solution.

Time (s)
Fig. 3. Time dependence of the potential of X52 steel electrode in the cathodic zone
at various DC current densities in the soil solution.
shows a bright, clear view, and corrosion products are not visible
on the steel surface. At the small DC current density of 0.1 A/m2,
the morphology of the steel electrode is similar to that in the
absence of DC. With the increase of the DC current density, the
steel surface becomes dark, and corrosion products can be seen
on the steel surface obviously. Moreover, as the DC current density
increases, the electrode surface turns darker and more corrosion
products are present on the surface. Generally, the DC application
induces uniform corrosion on the steel.
3.4. DC corrosion testing on the steel under CP at various DC current
densities
Fig. 8 shows the time dependence of the potential of X52 steel
electrode under CP potentials of 850 mV vs. CSE and 1000 mV
vs. CSE in the anodic zone at various DC densities in the soil solu-
tion, respectively. Prior to DC application, the steel is at the applied
CP potential. Similar to the potential records in the absence of CP in
Fig. 2, the potential of the steel cannot be maintained at the applied
CP potential, and is shifted towards the positive direction upon the
DC application. The steady-state DC-ON potential, i.e., the steel
potential under both CP and DC, is less negative with the increasing
DC current density. When the DC is stopped at the 2400th s, the
potential drops, but does not recover to the initial CP potential val-
ues. Generally, the new steady-state DC-OFF potential is more neg-
ative than the applied CP potential, especially at high DC current
densities, as shown in Fig. 8a and b.
Fig. 9 shows the time dependence of the potential of X52 steel
electrode at CP potentials of 850 mV vs. CSE and 1000 mV vs.
CSE in the cathodic zone at various DC densities in the soil solution,
respectively. It is seen that, similar to the potential records mea-
sured on the steel without CP application in Fig. 3, the potential
of the steel is shifted negatively under DC in the cathodic zone.
The DC-ON potential, i.e., the potential of the steel under both CP
and DC application, becomes more negative with the increasing
DC current density. When the DC is stopped, the potential of the
steel increases rapidly and gradually reaches a relatively stable
value, which is approximately close to the initially applied CP
potential.
Fig. 10 shows the potential of the X52 steel electrode at corro-
sion potential and varied CP potentials in the soil solution as a
function of DC current density. It is seen that, in the anodic zone,
the DC shifts the CP potential applied on the steel positively. At
individual DC current densities, the steel without CP has the least
negative DC-ON potential, and the steel applied at 1000 mV vs.
 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685 679

9.0 CSE has the most negative potential. In the cathodic zone, the DC is
(a) able to shift the CP potential applied on the steel more negatively
as the DC current density increases. At individual DC current den-
8.5 sities, the steel without CP application has the most negative
potential, and that applied at 1000 mV vs. CSE of CP has the least
8.0 negative potential.
Fig. 11 shows the solution pH in the anodic and cathodic zones
as a function of DC current density when the steel is at varied
pH

7.5 potentials. It is seen that, with the increase of the DC current den-
sity, the solution pH decreases and increases in the anodic and
cathodic zones, respectively. This is similar to those observed in
7.0 the absence of CP application in Fig. 5. At individual DC current
densities, the CP application could elevate the solution pH in both
6.5 the anodic and cathodic zones, compared to those measured in the
absence of CP in Fig. 5.
Fig. 12 shows the corrosion rate of the steel at corrosion poten-
6.0 tial and under CP potentials of 850 mV vs. CSE and 1000 mV vs.
0 2 4 6 8 10 12 CSE, respectively, determined by weight-loss testing after 48 h of
2
DC current density (A/m ) immersion in the soil solution as a function of DC current density.
Generally, the corrosion rate of the steel increases with the DC cur-
9.0 rent density. At individual DC current densities, the corrosion rate
(b) decreases if CP is applied on the steel. Moreover, the more negative
CP potential (i.e., 1000 mV vs. CSE) causes a smaller corrosion
rate. It is noted that, at a small DC current density of 1 A/m2, the
8.5 corrosion rate of the steel exceeds 1 mm/year even when the CP
of 850 mV vs. CSE is applied. The industry-acceptable corrosion
rate of pipelines is usually less than 0.1 mm/year [27]. Only when
the CP potential is shifted negatively to 1000 mV vs. CSE, the cor-
pH

8.0 rosion rate is negligible at 1 A/m2 of DC current density.

Fig. 13 shows the optical views of morphology of the steel that
is at corrosion potential and CP potentials of 850 mV vs. CSE and
1000 mV vs. CSE, respectively, after 48 h of testing in the soil
7.5 solution at various DC current densities. It is seen that the steel
experiences uniform corrosion under DC interference. Moreover,
the CP potential of 850 mV vs. CSE is not sufficient to protect
the steel from corrosion even at a small DC current density of 0.1
7.0 A/m2. When the CP potential is as negative as 1000 mV vs. CSE,
0 2 4 6 8 10 12
the steel is protected from corrosion at the DC current density up
2
DC current density (A/m ) to 1 A/m2. However, when the DC current density is increased to
2 A/m2, this CP level becomes not sufficient for corrosion protec-
Fig. 5. Solution pH in (a) anodic zone and (b) cathodic zone as a function of DC
tion. Obviously, in the presence of DC interference, the CP effec-
current density.
tiveness for corrosion protection is reduced remarkably.

4. Discussion
14
4.1. Accelerated corrosion of pipelines in the presence of DC
interference
12
Corrosion rate (mm/year)

This work demonstrates that the presence of DC stray current

10 can accelerate corrosion of pipelines in the extracted soil solution.
The electrochemical anodic and cathodic reactions during corro-
8 sion of pipeline steels in an aerobic soil solution include the steel
oxidation and the reduction of dissolved oxygen, which occur in
6 the anodic and cathodic zones, respectively.

Fe ! Fe2þ þ 2e ðAnodic reactionÞ ð1Þ

4

O2 þ 2H2 O þ 4e ! 4OH ðCathodic reactionÞ ð2Þ

2
Fig. 14a illustrates the mechanism of DC corrosion reactions
0 occurring at anode and cathode, respectively. Upon DC application,
0 2 4 6 8 10 12 the potentials of the steel in the anodic and cathodic zones are
shifted remarkably to the positive and negative directions, respec-
DC current density (A/m2)
tively, as seen in Figs. 2–4. For example, under the DC current den-
Fig. 6. Corrosion rate determined by weight-loss testing on the X52 steel coupon sity of 10 A/m2, the DC-ON potentials at the anode and cathode can
after 48 h of immersion in the soil solution as a function of DC current density. be up to about 400 mV vs. CSE and 1500 mV vs. CSE, respec-
680 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685

0 A/m2 0.1 A/m2

0.5 A/m2 1 A/m2

2 A/m2 5 A/m2

10 A/m2

Fig. 7. Optical views of the morphology of steel electrodes after 48 h of testing in the soil solution under various DC current densities.

tively, in the soil solution, while the corrosion potential of the X52
pipe steel is approximately 800 mV vs. CSE. Thus, the DC greatly
polarizes, either anodically or cathodically, the steel at the anodic
and cathodic zones, respectively. As a result, both the steel dissolu-
tion (corrosion) and the cathodic reduction of dissolved oxygen are
accelerated in the presence of DC interference. Moreover, new
steady-state potentials are established at the DC-ON levels, indi-
cating that the steel corrodes at a new state with a high corrosion
rate, as confirmed by the weight-loss measurements and corrosion
morphology observations in Figs. 6 and 7.
When the DC is removed, the potential is gradually recovered
back to the steady-state value which is generally identical to the
initial corrosion potential prior to DC application in the anodic
zone (Fig. 2). This indicates that the DC accelerated corrosion is
reversible under the present testing condition. Although the DC
application can cause the drop of solution pH in the anodic zone
(Fig. 5a), the pH values of the solution in the anodic zone before
and after application of the highest DC current density of 10 A/
m2 are still in the near-neutral pH range (changing from 7.7 to
6.4). The potential of the steel does not change greatly in the neu-
 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685 681

-0.2 -0.2
(a) 0.1 A/m2 (a)
0.5 A/m2
-0.4
-0.4 1 A/m2
10 A/m2
2 A/m2 Corrosion DC-ON DC-OFF

Potential (V vs. CSE)

Potential (V vs. CSE)

5 A/m2 -0.6 potential potential potential

-0.6 5 A/m2
10 A/m2
2 A/m2
1 A/m2 -0.8 0.1 A/m2
-0.8 0.5 A/m2
0.1 A/m2
-1.0
0.5 A/m2 0.1 A/m2
-1.0 1 A/m 2
0.5 A/m2
Corrosion DC-ON DC-OFF -1.2 2 A/m2 1 A/m2
potential potential potential 5 A/m2 2 A/m2
-1.2 -1.4 5 A/m2
10 A/m2
10 A/m2

-1.4 -1.6
0 500 1000 1500 2000 2500 3000 3500 4000 0 500 1000 1500 2000 2500 3000 3500 4000
Time (s) Time (s)

-0.2 -0.2
(b) 0.1 A/m2 (b)
0.5 A/m2 -0.4
-0.4 1 A/m2

Potential (V vs. CSE)

2 A/m2
Potential (V vs. CSE)

-0.6
5 A/m2
-0.6 10 A/m2 Corrosion DC-ON DC-OFF
10 A/m2
-0.8 potential potential potential
5 A/m2
2 A/m2
-0.8
1 A/m2 -1.0 0.1 A/m2
2 0.5 A/m2 1 A/m2
0.5 A/m 0.1 A/m2
-1.0 0.1 A/m2
-1.2 2 A/m 2 0.5 A/m2
5 A/m2
1 A/m2
Corrosion DC-ON DC-OFF 2 A/m2
-1.2 -1.4
potential potential potential 5 A/m2
10 A/m2
10 A/m2
-1.4 -1.6
0 500 1000 1500 2000 2500 3000 3500 4000 0 500 1000 1500 2000 2500 3000 3500 4000

Time (s) Time (s)

Fig. 9. Time dependence of the potential of X52 steel electrode at CP potentials of

Fig. 8. Time dependence of the potential of X52 steel electrode at CP potentials of
(a) 850 mV vs. CSE and (b) 1000 mV vs. CSE in the cathodic zone at various DC
(a) 850 mV vs. CSE and (b) 1000 mV vs. CSE in the anodic zone at various DC
densities in the soil solution.
densities in the soil solution.

tral pH range. If the DC current density is sufficiently large and the
anodic solution pH drops greatly to make the solution to be acidic,
it is expected that the steady-state DC-OFF potential of the steel
would be different from the original corrosion potential.
However, after the DC is stopped, the potential at the cathodic
zone does not return to the original corrosion potential. Generally,
the steel is at a new steady state with a less negative potential in
the cathodic zone, as seen in Fig. 3. Thus, the DC effect on the
cathodic zone is irreversible under the testing condition. The DC
drives the cathodic reduction of dissolved oxygen to produce
hydroxyl ions, elevating the solution pH. For example, the cathodic
solution pH increases from 7.7 to 8.6 upon the application of the
largest DC current density of 10 A/m2 in this work. The steel is
expected to be in a relatively more stable state in the new alkaline
environment, even after the DC is removed. Thus, the DC corrosion
of pipelines would cause local alkalization of the solution through
production of OH- by cathodic reaction. The degree of alkalization
depends on the value of DC current density.
4.2. Adverse effect of DC interference on CP performance
This work also demonstrates that, in the presence of DC, the
applied CP potential on the steel cannot be maintained, and is
shifted to the positive and negative directions in the anodic and
cathodic zones, respectively, as seen in Figs. 8–10. For example,
prior to DC application, the steel is at the CP potential of
850 mV vs. CSE and 1000 mV vs. CSE, which are both effective
for corrosion protection, as seen in Fig. 12. At the DC current den-
sity of 1 A/m2, the ‘‘true” potential of the steel is changed to about
750 mV vs. CSE and 880 mV vs. CSE, respectively, which are less
negative than the applied CP potential values. As a result, the CP at
850 mV vs. CSE is not able to protect the steel from corrosion, and
the corrosion rate of the steel increases to about 1 mm/year
(Fig. 12). When the DC current density is up to 10 A/m2, the CP
potential of 1000 mV vs. CSE losses its ability for corrosion pro-
tection because the ‘‘true” potential of the steel in the presence
of DC is about 650 mV vs. CSE only, where the corrosion rate is
up to approximately 4 mm/year. The steel under the applied CP
at 850 mV vs. CSE has a corrosion rate of about 9.3 mm/year at
the DC current density of 10 A/m2. Thus, the steel in the anodic
zone is not protected, at least not fully protected, by the CP from
corrosion in the presence of DC interference. The steel corrosion
at the anodic zone depends on both the DC current density and
the applied CP potential. As illustrated in Fig. 14b, when the
applied CP potential, such as 850 mV vs. CSE, is not sufficient to
offset the DC-induced anodic polarization, corrosion would occur
in the anodic zone. Unless the CP potential is sufficiently negative
682 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685

-0.3 8.5
(a) (a) Corrosion potential
-0.4 -850 mV vs. CSE
8.0 -1000 mV vs. CSE
-0.5
Potential (V vs. CSE)

-0.6
7.5

pH
-0.7

-0.8 7.0

-0.9 Corrosion potential

-850 mV vs. CSE 6.5
-1.0 -1000 mV vs. CSE

-1.1 6.0
0 2 4 6 8 10 12 0 2 4 6 8 10 12
DC current density (A/m2) DC current density (A/m ) 2

-0.7 10.5
(b) (b) Corrosion potential
Corrosion potential
-0.8 -850 mV vs. CSE
-850 mV vs. CSE 10.0 -1000 mV vs. CSE
-1000 mV vs. CSE
-0.9
Potential (V vs. CSE)

-1.0 9.5

-1.1
pH

9.0
-1.2

-1.3 8.5

-1.4
8.0
-1.5

-1.6 7.5
0 2 4 6 8 10 12 0 2 4 6 8 10 12
2 2
DC current density (A/m ) DC current density (A/m )

Fig. 10. Potential of the X52 steel electrode at corrosion potential and varied CP Fig. 11. Solution pH in the anodic (a) and cathodic (b) zones as a function of DC
potentials as a function of DC current density in the soil solution (a) anodic zone, (b) current density when the steel is at varied potentials.
cathodic zone.

14
(i.e., 1000 mV vs. CSE) so that the steel becomes cathodically Corrosion potential
polarized even in the presence of DC interference, the steel can 12 -850 mV vs. CSE
be protected from corrosion, as shown in Fig. 14c. -1000 mV vs. CSE
Corrosion rate (mm/year)

The DC effect on the potential in the cathodic zone is the nega-

10
tive shift of the steel potential with the increasing DC current den-
sity, which accelerates the cathodic reaction to produce hydroxyl
ions to elevate the solution pH. Compared to the environmental 8
alkalization in the absence of CP, the application of CP can further
enhance the solution alkalization. As seen in Fig. 11b, at CP poten- 6
tials of 850 mV vs. CSE and 1000 mV vs. CSE, the solution pH in
the cathodic zone can increase from 8.0 and 8.2 to about 9.2 and 4
10.0 at the DC current density of 10 A/m2, respectively.

2
4.3. Implications on pipeline integrity under DC interference

This work confirms that, in the presence of DC interference, the 0

0 2 4 6 8 10 12
pipeline integrity can be adversely affected by three aspects. First, 2
accelerating corrosion would occur in the anodic zone, which is DC current density (A/m )
usually associated with coating defects where the DC stray current
Fig. 12. Corrosion rate of the steel at corrosion potential and under CP potentials of
flows out of the pipeline steel into the conducting soil. The present 850 mV vs. CSE and 1000 mV vs. CSE, respectively, determined by weight-loss
work shows that, even at small DC current densities such as 5 and testing as a function of DC current density after 48 h of immersion in the soil
10 A/m2, the local corrosion rate of the steel can be 16 and 31 times solution.
 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685 683

0.1 A/m2

0.5 A/m2

1 A/m2

2 A/m2

5 A/m2

10 A/m2

Fig. 13. Optical views of morphology of the steel that is at corrosion potential (left) and CP potentials of 850 mV vs. CSE (middle) and 1000 mV vs. CSE (right), respectively,
after 48 h of testing in the soil solution at various DC current densities.

of that in the absence of DC interference. The accelerating corro- cause alkalinisation of the electrolyte remaining at the cathodic
sion at the anode could result in rapid perforation of the pipelines zone, i.e., the coating defect where the DC current flows into the
at the coating defect. Second, the presence of DC stray current can pipeline steel. Although corrosion would not occur locally, the
684 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685

Fig. 14. Schematic diagram of the mechanistic model to illustrate the electrochemical reactions occurring in the anodic and cathodic zones under DC interference (a) in the
absence of CP, (b) CP potential of 850 mV vs. CSE, (c) CP potential of 1000 mV vs. CSE and the DC current density is smaller than 1 A/m2.

highly alkaline electrolyte can degrade the polymeric coating, 5. Conclusions

causing disbonding of the coating from the steel substrate, i.e.,
the so-called ‘‘cathodic disbonding”. This problem is even worse
on CP-applied pipelines because, in addition to the DC stray
current, the CP could further elevate the local solution pH.
Finally, the DC is able to shift the applied CP potential, reducing
the CP effectiveness for corrosion protection. Thus, in the pres-
ence of DC interference, the CP criterion should be modified in
order to provide a full protection to the pipeline steel from
corrosion attack.
The DC stray current can greatly polarize, either anodically or
cathodically, the pipeline steel at the anodic and cathodic zones,
respectively, in an extracted soil solution. As a result, both the steel
corrosion and the cathodic reduction of dissolved oxygen are accel-
erated in the presence of DC interference. Under the present test-
ing condition, the DC accelerated steel corrosion at the anodic
zone is reversible, but its effect on the cathodic zone is irreversible
due to a strong solution alkalinisation is generated locally.
 S. Qian, Y.F. Cheng / Construction and Building Materials 148 (2017) 675–685
In the presence of DC interference, the applied CP potential on
the pipeline steel cannot be maintained, and is shifted to the pos-
itive and negative directions in the anodic and cathodic zones,
respectively. As a result, the steel in the anodic zone cannot be pro-
tected, at least not fully protected, by the CP from corrosion. The
solution alkalization in the cathodic zone would be further
enhanced by the combined effect of DC and CP on cathodic reduc-
tive reaction.
The DC stray current causes anodic and cathodic interferences
on the buried pipelines at the same time. For the anodic interfer-
ence, the DC current results in accelerated corrosion to the steel,
which presents an essential threat to the integrity of the pipeline.
For the cathodic interference, the dominant effect is the increased
pH of the local solution, which potentially causes disbonding of the
coating.
Acknowledgements
This work was supported by Eyes High Doctoral Recruitment
Scholarship at the University of Calgary – Canada.
References
[1] S. Papavinasam, R.W. Revie, Uhlig’s Corrosion Handbook, Wiley, New York, NY,
2000.
[2] B. Martin, A history of stray current corrosion, Corros. Mater. 31 (2006) 12–14.
[3] L. Schwalm, J. Sandor, Stray current-the major cause of underground plant
corrosion, Mater. Perf. 6 (1969) 31–36.
[4] W. Li, The monitor and control system of stray current corrosion in metro,
Urban Mass Transit. 4 (2003) 48–52.
[5] Y. Song, X. Wang, R. He, C. He, X. Wang, Status in research on stray-current
corrosion of buried steel pipelines, Corros. Prot. 8 (2009) 003.
[6] A. Fu, Y. Cheng, Effects of alternating current on corrosion of a coated pipeline
steel in a chloride-containing carbonate/bicarbonate solution, Corros. Sci. 52
(2010) 612–619.
[7] D. Kuang, Y. Cheng, Understand the AC induced pitting corrosion on pipelines
in both high pH and neutral pH carbonate/bicarbonate solutions, Corros. Sci.
85 (2014) 304–310.
685
[8] A. Pourbaix, P. Carpentiers, R. Gregoor, Detection and assessment of alternating
current corrosion, Mater. Perf. 39 (2000) 34–37.
[9] A. Brenna, S. Beretta, F. Bolzoni, M. Pedeferri, M. Ormellese, Effects of AC-
interference on chloride-induced corrosion of reinforced concrete, Const. Build.
Mater. 137 (2017) 76–84.
[10] L.V. Nielsen, Role of Alkalization in AC Induced Corrosion of Pipelines and
Consequences Hereof in Relation to CP Requirements, Corrosion 2005, NACE
International, Houston, USA, 2005.
[11] H. Xiao, S.B. Lalvani, A linear model of alternating voltage-induced corrosion, J.
Electrochem. Soc. 155 (2008) C69–C74.
[12] L. Xu, X. Su, Z. Yin, Y. Tang, Y. Cheng, Development of a real-time AC/DC data
acquisition technique for studies of AC corrosion of pipelines, Corros. Sci. 61
(2012) 215–223.
[13] L. Xu, X. Su, Y. Cheng, Effect of alternating current on cathodic protection on
pipelines, Corros. Sci. 66 (2013) 263–268.
[14] E. Ghanbari, M. Iannuzzi, R. Lillard, The mechanism of alternating current
corrosion of API grade X65 pipeline steel, Corrosion 72 (2016) 1196–1210.
[15] L. Bertolini, M. Carsana, P. Pedeferri, Corrosion behaviour of steel in concrete in
the presence of stray current, Corros. Sci. 49 (2007) 1056–1068.
[16] F. Brichau, J. Deconinck, T. Driesens, Modeling of underground cathodic
protection stray currents, Corrosion 52 (1996) 480–488.
[17] C. Andrade, I. Martinez, M. Castellote, Feasibility of determining corrosion
rates by means of stray current-induced polarisation, J. Appl. Electrochem. 38
(2008) 1467–1476.
[18] D. McIntosh, Grounding where corrosion protection is required, IEEE Trans.
Indu. Appl. (1982) 600–607.
[19] D. Jones, Effect of alternating current on corrosion of low alloy and carbon
steels, Corrosion 34 (1978) 428–433.
[20] P.M. Scott, Environmentally induced cracking, Construct. Build. Mater. 4
(1990) 98–106.
[21] I. Metwally, H. Al-Mandhari, A. Gastli, Z. Nadir, Factors affecting cathodic-
protection interference, Eng. Ana. Boundary Elements 31 (2007) 485–493.
[22] I. Metwally, H. Al-Mandhari, A. Gastli, A. Al-Bimani, Stray currents of ESP well
casings, Eng. Ana. Boundary Elements 32 (2008) 32–40.
[23] S. Srikanth, T. Sankaranarayanan, K. Gopalakrishna, B. Narasimhan, T. Das, S.K.
Das, Corrosion in a buried pressurised water pipeline, Eng. Failure Ana. 12
(2005) 634–651.
_
[24] K. Zakowski, W. Sokólski, Numeric method of IR elimination from
measurement results of potential of underground pipelines in the zone of
interaction of stray currents, Corros. Sci. 41 (1999) 2213–2222.
[25] NACE Standard RP0169-2002 Control of external corrosion on underground or
submerged metallic piping systems, NACE, 2002.
[26] ASTM Standard Practice for Preparing, Cleaning, and Evaluating Corrosion Test
Specimens, ASTM International, 2011.
[27] J.R. Davis, Corrosion: Understanding the Basics, ASM International, 2000.